On Sat, Jun 23, 2018 at 6:19 AM, Jens Axboe [off-list ref] wrote:

On 6/22/18 4:12 PM, Ming Lei wrote:

quoted

SCSI probing may synchronously create and destroy a lot of request_queues
for non-existent devices. Any synchronize_rcu() in queue creation or
destroy path may introduce long latency during booting, see detailed
description in comment of blk_register_queue().

This patch removes two synchronize_rcu() inside blk_cleanup_queue()
for this case:

1) commit c2856ae2f315d75(blk-mq: quiesce queue before freeing queue)
need synchronize_rcu() for implementing blk_mq_quiesce_queue(), but
when queue isn't initialized, it isn't necessary to do that since
only pass-through requests are involved, no original issue in
scsi_execute() at all.

2) when only one request queue is attached to tags, no necessary to
call synchronize_rcu() too.

Without this patch, it may take more 20+ seconds for virtio-scsi to
complete disk probe. With this patch, the time becomes less than 100ms.

Reported-by: Andrew Jones <redacted>
Cc: Andrew Jones <redacted>
Cc: linux-scsi@vger.kernel.org
Cc: "Martin K. Petersen" <martin.petersen@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <redacted>
---
 block/blk-core.c | 8 ++++++--
 block/blk-mq.c   | 5 ++++-
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index cf0ee764b908..f0129e20b773 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c

@@ -766,9 +766,13 @@ void blk_cleanup_queue(struct request_queue *q)
       * make sure all in-progress dispatch are completed because
       * blk_freeze_queue() can only complete all requests, and
       * dispatch may still be in-progress since we dispatch requests
-      * from more than one contexts
+      * from more than one contexts.
+      *
+      * No need to quiesce queue if it isn't initialized yet since
+      * blk_freeze_queue() should be enough for cases of passthrough
+      * request.
       */
-     if (q->mq_ops)
+     if (q->mq_ops && blk_queue_init_done(q))
              blk_mq_quiesce_queue(q);

      /* for synchronous bio-based driver finish in-flight integrity i/o */

diff --git a/block/blk-mq.c b/block/blk-mq.c
index 70c65bb6c013..8a6771ac0adb 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c

@@ -2351,6 +2351,7 @@ static void blk_mq_update_tag_set_depth(struct blk_mq_tag_set *set,
 static void blk_mq_del_queue_tag_set(struct request_queue *q)
 {
      struct blk_mq_tag_set *set = q->tag_set;
+     bool shared;

      mutex_lock(&set->tag_list_lock);
      list_del_rcu(&q->tag_set_list);

@@ -2360,8 +2361,10 @@ static void blk_mq_del_queue_tag_set(struct request_queue *q)
              /* update existing queue */
              blk_mq_update_tag_set_depth(set, false);
      }
+     shared = set->flags & BLK_MQ_F_TAG_SHARED;
      mutex_unlock(&set->tag_list_lock);
-     synchronize_rcu();
+     if (shared)
+             synchronize_rcu();

Shouldn't this be set if it _was_ shared as well, not just if it's
still shared?

Yes, it need to be done for _was_ shared. But for the usual single lun case,
the shared can be set for all the following probe suppose lun 0 is the real one,
then this simple trick can't work any more.

Looks blk_queue_init_done() still need to be check here.


thanks,
Ming Lei